Malaria control and elimination in areas of high transmission in sub-Saharan Africa present a significant challenge to global health. A large fraction of the population across all ages in these areas harbor Plasmodium falciparum without clinical manifestations, providing a vast reservoir of infection for transmission. This asymptomatic reservoir is sustained by the enormous antigenic diversity of the parasite. Thus, the challenge for hyperendemic regions requires that the malaria field comes to terms with such diversity, studying it as a complex adaptive system. This study addresses the two-way interaction between epidemiology and P. falciparum diversity from the perspective of the multigene and highly recombinant family known as var, which encodes for the major antigen of the blood stage of infection. This project combines theory with field and laboratory work lo generate new understanding of the diverse transmission reservoir of P. falciparum and its resilience to elimination. The first aim is the longitudinal deep sampling of this reservoir in the Bongo District, Ghana following two control interventions (i.e. indoor residual spraying and seasonal malaria chemoprevention). On the basis of age-stratified serial cross-sectional data, this study will assess how informative the var system is to monitor this reservoir under conditions of the control interventions, compared to traditional malariometric indices and neutral molecular markers. The second aim develops strain theory based on a stochastic agent-based model that tracks the history of infection of each host and the evolutionary change of the parasite. Pathogen population structure over time and responses of the var system to intervention are investigated in ways that inform both molecular data analyses and control efforts. The third aim develops a transmission model of intermediate complexity that can be fitted to routine epidemiological data but still incorporates the major effects of parasite diversity on epidemiology. In particular, the existence of a threshold in transmission intensity that impacts parasite antigenic diversity will be investigated. Contributions include computational models at the interface of epidemiology and evolution, and network analyses of population genetic structure applicable to other multigene families of P. falciparum, as well as to other pathogens whose immune escape relies on highly-recombinant gene families.